Extraction of urban features with knowledge based engineering and local maxima filtering
The aim of this study is creating a tool, which has the ability to incorporate different types of remotely sensed data in such a way, that not only buildings but also tree tops can be extracted from it. Another aim is to extract the desired results with an automated approach. Numerous algorithms are created with the help of the programming language Python. The first step is using a knowledge based engineering to identify buildings and tree tops. The next step is to enhance the data with the help of binary filters, which are used to reduce the noise and refill missing data points. To extract the building the raster data are converted into vectors and then structures are removed, which don't fulfil user set criteria. The tree tops are located by a local maxima filtering and the final product is created by applying rules for the tree tops. It can be demonstrated, that the success mainly depends on the careful selected values for the user set parameters.